REVIEW Open Access

Techniques and future perspectives for theprevention and treatment of endoleaksafter endovascular repair of abdominalaortic aneurysmsGianluigi Orgera1, Marcello Andrea Tipaldi1, Florindo Laurino1, Pierleone Lucatelli2, Alberto Rebonato3,Ioannis Paraskevopoulos4, Michele Rossi1 and Miltiadis Krokidis5*

Abstract

The presence of endoleaks remains one of the main drawbacks of endovascular repair of abdominal aortic aneurysmsleading to the increase of the size of the aneurysmal sac and in most of the cases to repeated interventions. A varietyof devices and percutaneous techniques have been developed so far to prevent and treat this phenomenon, includingsealing of the aneurysmal sac, endovascular embolisation, and direct sac puncture. The aim of this review is to analysethe indications, the effectiveness, and the future perspectives for the prevention and treatment of endoleaks afterendovascular repair of abdominal aortic aneurysms.

Keywords: Aneurysm, Endoleak, Aorta

Key points

� The detection rate of endoleaks depends on theimaging modalities used

� Only a small percentage of endoleaks will requirere-intervention

� Treatment may include both endovascular orpercutaneous route

IntroductionThe endovascular aortic repair (EVAR) of abdominalaortic aneurysms was first described nearly three decadesago and has offered a crucial management shift of patientswith aortic disease, particularly when open repair was notan option [1–3].A variety of EVAR devices were developed over the

years offering a range of outcomes. There has been asubstantial evolution in design and technology, from theinitial tube grafts to the custom-made fenestrated andbranched devices that are used today. EVAR has offered

some benefits over the traditional open surgical repair;however, there is a cost to pay and this is mainly theneed of closer patient follow-up and sometimes thenecessity of re-interventions [4–9]. Follow-up is requiredto assess growth of the aneurysm sac, device migration,blockage, or infection. The most common reason for re-intervention is the increase of the aneurysmal sac duepersistence of flow, a phenomenon otherwise known as“endoleak” [4–9]. The purpose of this review article is toillustrate the various types of endoleaks and to describewhat the current status of percutaneous management is.

Classification of endoleaksEndoleaks are classified into five types (I–V). Type Ioccurs due to incomplete proximal (Ia) or distal (Ib) seal.This could be due to either inappropriate device selection,incorrect graft deployment, or disease progression [10].Type Ia may also appear when chimneys are used and areknown as “gutter endoleaks” [11, 12]. Type II occurs dueto sac centripetal reperfusion via side branches (lumbararteries, inferior mesenteric artery, accessory renal arte-ries) with inverted flow. Type III is a result of dislodge-ment of the various graft components. Type IV occurs dueto increased porosity of the graft material. Type V (also

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made.

* Correspondence: mkrokidis@hotmail.com5Department of Radiology, Cambridge University Hospitals NHS FoundationTrust, Hills Road, Cambridge CB2 0QQ, UKFull list of author information is available at the end of the article

Insights into ImagingOrgera et al. Insights into Imaging (2019) 10:91 https://doi.org/10.1186/s13244-019-0774-y

known as “endotension”) is the type of endoleak that can-not be classified in any of the other categories.

IncidenceIt is not easy to define the precise incidence of endoleaksfrom the existing data in the literature. The evidence-based practice centre of the Mayo Clinic has publishedin 2017 a systematic review and meta-regression analysisevaluating surveillance outcomes after EVAR for AAAincluding 6 meta-analyses and 52 observational studies[13]. The authors confirmed that an endoleak incidencerate is subject to the type of imaging modality used fortheir detection and reported that a combined approachof DUS, non-contrast-enhanced CT, and MRI offeredthe highest endoleak detection rate at 12, 24, 36, and 48months of 35%, 46%, 51%, and 92%, respectively. At 60months, the highest detection rate (91%) was observedusing a combined approach of DUS, CTA, and MRI.However, most of the centres adopt the use of CTA withdelayed images as the “gold standard” as it is the mostcost-effective single modality for endoleak detection.Operator experience and appropriate sizing decreased

the incidence of type I (both a and b) endoleaks with theuse of conventional devices over the years. However, theincidence of type Ia is still high when chimneys are used,reaching sometimes even 30% [14]. Type II endoleaks re-main stable over the years in terms of incidence andhave been reported around 10% for emergency EVARsand around 20% after elective repairs [15]. Types III–Vendoleaks have been reported with lower incidence, inthe region of 1–3% [16].

Symptoms and managementUnderstanding the aforementioned different pathophysi-ology mechanisms that lead to the various types of endo-leaks permits to distinguish the high- and low-flow nature,which impacts on the requirement or not of repeated inter-vention and correction. In particular, high-flow endoleakscould lead to the rupture of the aneurysmal sac and maybe associated with symptoms like lumbar pain, or evenhypotension and tachycardia due to hypovolemic shock[17]. On the other hand, type II endoleaks due to theirpathophysiology mechanism of onset are classified as lowflow. This is because the sac reperfusion occurs via col-lateral circulation arising with inverted flow from theperiphery towards the aneurysm sac. In these cases, thesac is exposed to a low-pressure flow that may lead tosac enlargement over time. The evolution can be slowand the endoleak can either be managed conservatively[18, 19] or lead to sac expansion over 5 mm in a yearin which case treatment will be deemed necessary[20–23]. According to the type of endoleak, differentapproaches may be followed [24].

Type I endoleaksType Ia may have an immediate onset after graft deploy-ment that offers the option to treat it in the same ses-sion. Type Ib usually has a late onset that would requirelate re-intervention only in case of sac growth.

Intra-procedure detectionIf a type Ia is detected immediately after graft deploy-ment, the first approach is to expand further the neck ofthe graft with a moulding balloon. If the effect of theballoon is not satisfactory or if the type Ia is a result of alow graft deployment then the deployment of a cuffneeds to be considered (Fig. 1). Cuffs are available formost of the devices; however, if a cuff is not availableand the moulding balloon does not appear to work, thenthe use of a bare stent (Palmaz stent, Cordis Inc) thatmay expand to the desired diameter needs to be consid-ered. The long-term results of the use of Palmaz stentsin treating intra-procedural type Ia endoleak, in patientswhom proximal stent graft cuff implantation was notfeasible, has been published by Abdulrasak et al. [25]They reported that between 1998 and 2012, 125 patientswere treated endovascularly in both elective and emer-gency settings (83 elective and 42 emergencies) demon-strating a primary and assisted freedom from type Iaendoleak at 5 years of 84 ± 4% and 89 ± 3%, respectively(elective vs emergency cases). Recently, endoanchorshave been introduced that may be used to fix further thegraft to the aortic wall particularly when the type Ia isdue to a short neck [26] (Fig. 2).

Late detectionIf the type Ia is detected in one of the follow-up scans,then all the abovementioned approaches may be equallyused according to the anatomy and the operator’s prefer-ences. However, if there is disease progression with neckexpansion and no suitable anatomy for a cuff, a Palmazstent or the use of endoanchors then extension of the graftin the suprarenal and visceral segment is required withthe use of more complex chimney/periscope techniquesor the use of a fenestrated or branched cuff. The use ofchimneys and periscopes are rescuing endovascular tech-niques that employ the shelf stent grafts to extend theproximal or distal landing zones. Stent grafts are deployedparallel to an aortic extension in order to preserve flowwithin the aortic visceral vessel branches. Chimneys areperformed to divert flow from the aortic lumen towardsthe branches in a standard anatomy (proximally-distally),whereas periscopes divert flow from the aortic lumen to-wards the branches in a reversed direction. Montelione etal. [27] reported 12 years experience in 24 patients pre-senting a type Ia endoleak after a previous endovascularaneurysm repair (EVAR) treated with chimney and/orperiscope grafts. They demonstrated a technical success of

Orgera et al. Insights into Imaging (2019) 10:91 Page 2 of 10

96%, with effective intraoperative revascularisation of alltarget vessels; moreover an estimated survival at 12, 24,and 36months of 83% and estimated snorkel/chimney pa-tency at the same intervals of 94%. The authors demon-strated aneurysm sac shrinkage during a mean follow-upof 23.4 ± 29months. Mean maximal aneurysm sac diam-eter decreased from 88 ± 26 to 85 ± 33mm (p = 0.49) overthe course of the follow-up. Aneurysm sac diameterremained stable or decreased in 21 (87%) patients; theother 3 had sac diameter increases > 5mm, one of whichwas related to a recurrent type Ia endoleak.In the case of type Ib endoleaks, embolisation of the

ipsilateral internal iliac artery and distal graft extensionis usually the most straightforward solution [28, 29].

Type II endoleaksPreventionSeveral studies have examined the effectiveness of pre-operative embolisation of branch vessels for the preventionof type II endoleaks [30–32]. Alerci et al. [31] evaluated theembolisation of large lumbar arteries prior to EVAR on124 patients. The rate of type II endoleaks was significantlylower in the embolisation group (3.6% vs 47.8%, p < 0.001)after a mean clinical follow-up of 60.5 ± 34.1months(range 1–144). Piazza et al. [32] reported similar results

and also demonstrated that patients who underwent pre-operative embolisation experience faster sac shrinkage andthat the only independent predictor of a type II endoleakoccurrence is preoperative aneurysm sac volume > 125cm3. A meta-analysis by Biancari et al. [33] demonstrated apooled rate of type II endoleak after IMA embolisation of19.9% versus 41.4% in patients without IMA embolisation.However, despite this different prevalence of endoleaks,the authors conclude that since treatment for type II endo-leaks is needed in less than 20% of cases and this complica-tion can be treated successfully in 60–70% of casesresulting in an aneurysm rupture risk of 0.9%, embolisationof patent IMA may be not of overall benefit to patientsundergoing EVAR. Nevertheless, there are groups that stillsuggest that IMA embolisation may be a preventive meas-ure particularly for arteries with a diameter > 3mm [34].Furthermore, the role of intra-procedural embolisation

of the aneurysmal sac with thrombin and gelfoam slurryhas been investigated, in limited series, demonstrating atrend towards lower rate of type II endoleaks when com-pared with the literature median rate [35]. Zanchetta et al.[36], in a prospective, nonrandomised pilot study used fi-brin glue aneurysm sac embolisation at the time of EVARin 84 patients and demonstrated a low rate of delayed typeII endoleak and a statistically significant decrease in the

Fig. 1 a Coronary reconstruction of CTA scan confirming deployment of the graft in a caudal position led to sac expansion in the follow-upperiod. b Angiogram confirmed the low graft position. c A cuff was deployed in an immediate infrarenal position to prevent any furthersac expansion

Fig. 2 Endoanchors were employed to fix the proximal graft given the short (< 10mm) neck. a, b Fluoroscopic picture showing the delivery ofthe anchors. c Angiogram confirming the good apposition of the graft after the deployment of four endoanchors

Orgera et al. Insights into Imaging (2019) 10:91 Page 3 of 10

maximum transverse aneurysm diameter at follow-up.Muthu et al. [37] in a study performed in 2007 wherepatients that received IMA embolisation combined withintraprocedural thrombin injection in 69 consecutivepatients that underwent elective EVAR showed a trend ofendoleak reduction; however, no statistically significantdifference was reached (26% compared with 14%) and theauthors concluded that ongoing research into means toprevent lumbar endoleaks is required. Even though thereis a variety of studies available, there is still not enoughevidence to support routine embolisation of the IMA or thelumbar arteries or intraprocedural thrombin injectionprior to routine EVAR and this remains an operator’spreference [38].To avoid the doubt of preoperative embolisation for

the prevention of type II endoleaks, the Nellix EVASdevice (Endologics) was developed aiming to “seal” thesac instead of blocking the centripetal flow. The Nellixsystem comprises two 10-mm balloon expandablechromium-cobalt stent grafts that are inserted into theaorta in a “double-barrel” conformation. A bag isattached to each of the stent grafts and is filled with apolymer during insertion in order to conform accord-ing to the anatomical shape of the flow component of

the aneurysmal sac [39]. By filling the aortic lumen,the endobags eliminate the space in the sac and limitthe possibility of flow towards it. Therefore, when Nel-lix was launched few years ago, there was a high ex-pectation on limiting the rate of type II endoleaks.Nevertheless, the device did not perform at the stan-dards that it was initially expected, probably becausein most of the cases it was used outside the instruc-tions for use (IFU) [40–43]. Specifically, the IFUinstructed aortic proximal neck diameter range of 18to 28 mm, minimum aortic proximal neck length ≥ 10mm, proximal aortic neck angulation of ≤ 60°, aorticaneurysm with a blood lumen diameter ≤ 70 mm, ratioof maximum aortic aneurysm diameter to maximumaortic blood lumen diameter < 1.4, and distal iliac ar-tery seal zone with length of ≥10 mm and diameterrange of 9 to 25 mm. When the device was used withinIFU, like in the series of Carpenter et al. [44], it per-formed much better. However, given the erroneoususe, type Ia endoleaks were developed that were im-possible to control, leading to graft separation and sacexpansion (Fig. 3). Most of the devices were explantedand the graft lost the European Conformity mark inJanuary 2019.

Fig. 3 a Coronary reconstruction of a CT scan showing the satisfactory deployment of a Nellix device with lack of separation of the grafts. bTransverse CT scan confirming the sac size after deployment. c, d Follow-up CT 2 years later shows separation of the grafts and sac expansion.This is a result of a subtle type Ia endoleak between the two grafts.

Orgera et al. Insights into Imaging (2019) 10:91 Page 4 of 10

Endovascular treatmentRegarding the endovascular treatment of type II endoleaks,the aim is to reach the aneurysmal sac and block the feed-ing vessels with embolisation material in order to controlthe sac growth [45]. The approach differs according to thecollateral pathways involved in the sac reperfusion. Thefocus of treatment is to seal all access to the sac and if pos-sible to directly embolise the sac to avoid recurrence. If theendoleak is supplied by the IMA, a retrograde approachfrom the superior mesenteric artery and the Riolan arcadewill be required (Fig. 4) [46]. In the case of supply via thelumbar arteries, catheterisation via the ileo-lumbar anasto-moses that take origin from the ipsilateral internal iliacartery should be performed. The enlargement of this colla-teral pathway may allow the navigation with a microcath-eter permitting super selective catheterisation of thefeeding lumbar arteries. Ribè et al. [47] reported a technicalsuccess of 100% and no endoleak recurrence via the treatedcollateral pathways till 19months of follow-up employingan Onyx liquid embolic agent.

Percutaneous treatmentAnother way of treating the type II endoleaks is via directpercutaneous sac access under ultrasound (US), computedtomography (CT), or digital subtraction angiography(DSA) guidance (Fig. 5).

Baum et al. [48] described the first experience withtranslumbar embolisation in literature of type II endoleaksafter endovascular repair of abdominal aortic aneurysms.Translumbar paraspinal left-side access with a patient inthe prone position is the most common generally per-formed approach to avoid the inferior vena cava (IVC)while the choice of transabdominal is preferred when anendoleak sac is located or extended anteriorly from theaneurysm sac. Stavropoulos et al. [49] described a proneright-sided percutaneous transcaval approach in order tomanage either position of the endoleak or interposedbowel and organs.The direct access should be performed with a micro

puncture access set and guided by ultrasound, CT, orDSA depending on either the anatomy or the operator’spreference. Van Bindsbergen et al. [50] described a tech-nique for real-time three-dimensional needle guidanceusing cone-beam CT co-registered to fluoroscopy, whichcan potentially facilitate treatment of small a nidusdecreasing radiation and contrast doses compared withtraditional methods. Once the nidus is accessed, the nee-dle is exchanged over a guidewire for a vascular sheathand a 4Fr or 5Fr selective catheter. Contrast media isinjected to obtain an endoleak-o-gram to define the sizeof the nidus and the location of the inflow and outflowarteries. Embolisation is performed until the nidus is

Fig. 4 Percutaneous embolisation for type II endoleak. a Delayed CTA reveals the presence a type II endoleak (arrow). b .Angiogram confirmingthe access via the Riolan arcade to the IMA. c Embolisation with Onyx and (d) satisfactory angiographic result

Orgera et al. Insights into Imaging (2019) 10:91 Page 5 of 10

occluded. The feeding vessels should also be embolisedif feasible, but this may be problematic particularly inthe case of lumbar arteries due to their unfavourableanatomy as U-turn shape to extend posteriorly [51].Furthermore, an alternative access route to reach the

aneurysmal sac has been reported via the inferior venacava as mentioned [52–54]. In these cases, theaneurysmal sac is accessed through the inferior venacava thanks to the employment of a transhepatic shuntaccess needle set in order to puncture the sac. Then, themicrocatheter is advanced within the sac and the embol-isation performed. Scali et al. [52] reported experienceon transcaval embolisation with coil placement and se-lective thrombin injection in six patients with 100%technical success and no postoperative complications. Inone patient, repeated treatment was required at follow-up of 8.1 months. Giles et al. [54] reported a larger ex-perience in 26 patents who underwent transcaval coilembolisation of the aneurysm sac at a mean of 4.2 ± 4years after initial endovascular aneurysm repair. Therewere no procedural adverse events and re-interventionwas required in five cases.A new very interesting percutaneous approach has

been reported by Ogawa et al. [55] who performed atranspedicular direct puncture using an isocentre punc-ture method: an isocentre marker was placed at a sitecorresponding to the aneurysm sac on fluoroscopy intwo directions (frontal and lateral views); then, a verteb-roplasty needle was inserted tangentially to the markerunder fluoroscopy and advanced to the anterior wall ofthe vertebral body. Finally a 20 cm-length, 20G needlewas inserted through the outer needle of the 13G needleand advanced to the marker.The embolic agent options include mechanical devices

and liquids, used in combination or alone, with a choiceof an embolic material tailored to the patient’s anatomyand operator preference. There are no conclusive dataon the advantages or disadvantages of the different em-bolic agents to date, even if high rates of endoleak recur-rence (50 %) have been reported in small case series

with the use of thrombin as the predominant embolicagent [54], and it seems its use should be avoided.Some complications of the direct approach have been

reported to date, the most clinically significant is a pul-monary embolus secondary to extravasation of the gluein the IVC, stent puncture leading to the developmentof a type III endoleak [54] and bowel ischemia [52].

Comparison of endovascular and percutaneousmanagementA debate in the literature exists between the use ofDSPE and transarterial embolisation (TAE) but there isno consensus on which is the most effective or first-lineapproach for type II endoleaks. Even though there areseveral retrospective series of patients comparing thetwo techniques, there is no prospective randomisedcomparison of TAE versus DSPE. Baum et al. [48] pub-lished the first study in 2002 where 20 patients under-went TAE of the IMA and 13 patients underwent directtranslumbar embolisation. Over the follow-up period of254 days, 16 out of the 20 transarterial embolisation, pa-tients required re-intervention due to recanalisation ofthe initially embolised vessels. Most of the DSPE proce-dures however were successful with only one recurrence.The authors report a 92% success rate after DSPE usingcoils, compared with an 80% failure rate of transarterialcoil embolisation of the feeding vessels. They concludedthat DSPE should be the technique of choice for type IIendoleaks. This was a milestone study that triggered thedevelopment of further percutaneous techniques.Sidloff et al. [18] in a systematic review reported an

overall failure rate of 37.5 % for TAE compared with a19% for DSPE; however, it is important to highlight thatin many studies, the translumbar approach represents a“bail-out” of the transarterial one and the results wouldprobably have been different if this was the initial ap-proach. On that note, Uthoff et al. [56] conducted a sin-gle-centre retrospective analysis of 19 type II endoleakstreated via translumbar approach, and they demon-strated an initial technical success rate of 88%; however,

Fig. 5 a Direct puncture of the sac under CT and fluoroscopic guidance. Angiogram confirms the presence of the small nidus and the feedingvessel. b CT scan during embolisation with Onyx

Orgera et al. Insights into Imaging (2019) 10:91 Page 6 of 10

in half of these patients recurrence occurred after 39months of follow-up and in two-thirds of them a second-ary procedure was necessary. In a paper of 2016, Yang etal. [57] reported on twenty-three type II endoleak patientssimilar sac occlusion effectiveness, between direct sacpuncture and transarterial embolisation. The median fol-low-up was 21.8months. Direct access was considered asthe preferred approach due to shorter fluoroscopic andprocedural times. The studies from the literature compar-ing DSPE and TAE are shown in Table 1.

Types III, IV, and V endoleaksType IIIChaikof et al. [58] classified first type III endoleaks in2002 in “Reporting standards for endovascular aorticaneurysm repair”. They may occur either due to discon-nection of components of the modular endograft system(IIIa) or a defect in the stent-covering graft fabric (IIIb).Type III b endoleaks are further stratified with respectto the extent of fabric disruption as major (> 2 mm) orminor (< 2 mm). Incidence of type III endoleaks hasbeen reported from 1 to 11% [59]. Also based on thecomplexity of the anatomy, type IIIa endoleak could beclassified as simple and complex. Improper seal,inadequate overlap of modular components, and distalcomponent migration are considered as simple type IIIendoleaks, whereas major component dislocation includ-ing total mal-alignment results in complex type IIIaendoleaks. With the advent of more complex fenestratedEVAR and TEVAR procedures, the total number ofjunctions in endografts has increased with the possibilityof a corresponding increase in the incidence of type IIIendoleaks. Type III endoleaks arising from side branchesare a special concern after fenestrated endografting, withreported rates of 0 to 6.8% [60].Treatment consists in either realigning the dislodged

stent graft parts or advancing another covered stent in

the dislodged branches/ fenestrations in the b-EVAR/ f-EVAR case.

Type IVIn patients who underwent EVAR repair, type IV endo-leaks are very rare [61, 62]. Espinosa et al. reported aprevalence of this type of endoleak as 0.3% [63]. Forbeset al. reported the conservative treatment as sufficient,in the case of type IV endoleak. Furthermore, in longerfollow-up observation, a type IV endoleak was not thecause of the aneurysm re-supply in open surgery [64]and some clinicians think that type IV endoleaks shouldbe classified as a type V endoleak.Type IV endoleaks are even more rare with the new-

generation stent grafts. In such cases, usually, the bestmanagement is surgical conversion with graft explant-ation [65, 66].

Type V endoleaksType V endoleaks are a result of “endotension”. With thisterm is defined the aneurysm sac growth without any de-tectable endoleak and is a result of increased pressurewithin the aneurysm sac (Fig. 6) [67]. This may be due tosuch slow blood flow that it is below the sensitivity limitsfor detection on current imaging methods. This kind ofendoleak seems to be more common with expanded poly-tetrafluoroethylene (ePTFE) fabric grafts rather than poly-ester covered ones. New-generation ePTFE grafts includea second layer of low permeability ePTFE to decrease thisrisk [68]. No studies have shown an increased risk ofaneurysm rupture among patients with endotension [69,70]. However, the current view regarding type V manage-ment is that when sac growth of more than > 8mmoccurs, then some form of re-intervention is required [71,72]. Nevertheless, further studies are required to demystifythis phenomenon and condition and delineate the appro-priate management.

Table 1 Studies in the literature that compare the direct sac with the endovascular approach for the treatment of Type II endoleaks.Technical success*: immediate exclusion of the sac at the first control. Clinical success**: freedom from endoleak recurrence at thefollow-up. DSPE, direct sac puncture embolisation; TAE, transarterial embolisation

Study No. Mean follow-up time Technical success* (%) Clinical success** (%)

Baum et al. [48] 20 TAE 254 days 90 20

13 DSPE 254 days 100 92

Stavropoulos et al. [49] 23 TAE 17.3 months 95.7 78.3

62 DSPE 20.2 months 100 72.6

Nevala et al. [74] 10 TAE 4.5 ± 2.3 years 40 20

4 DSPE 4.5 ± 2.3 years 100 75

Massis et al. [75] 65 TAE 15 weeks 58 76

36 DSPE 15 weeks 100 59

Yang et al. [57] 23 DSPE 21.8 months 100 64

81 57

Orgera et al. Insights into Imaging (2019) 10:91 Page 7 of 10

Future perspectivesConsidering the impact that endoleak prevention andtreatment have for health economics, there is continuousresearch on the field with a prediction for an exponentialincrease in the next 5 years. The main areas that will bedeveloped are the following:

– Imaging modalities, mainly for the early detectionof and characterisation endoleaks, aiming forradiation-free modalities like contrast-enhancedultrasound (CEUS) and magnetic resonance imaging.The version 1.3 of the study on the early detectionof endoleaks with CEUS (NCT02688751) is recentlycompleted. The primary outcome is to assess theability of CEUS to detect type I/III endoleaks onCEUS as defined by presence/absence on time-resolved CTA. The secondary outcome is thedetection of type II endoleaks and the ability ofCEUS to predict the likelihood of a secondaryintervention. The study has also assessed healthcarecosts related to each imaging modality, consideringthat EVAR follow-up carries an important economicimpact. The results have not been made public yet.

Radiation reduction can also be achieved with dual-en-ergy CT that acquires two different photon spectra in asingle acquisition. It can be used to detect endoleakswith good accuracy and at a reduced radiation exposureand some preliminary data is already available [73].

– Biomarkers that would predict the aneurysmevolution. The best example is the matrixmetalloproteinase (MMP) activity that has beenassociated with the process of aneurysmdevelopment. In essence, if there is a lack of balancebetween the MMP and its inhibitors, degenerationof the aortic wall is induced. It was previouslyproved that the serum level of MMP-9 issignificantly higher in patients with abdominal aorticaneurysm and in patients with inadequate aneurysmexclusion after EVAR. A multicentre trial of serumlevels of MMP-9 as a biomarker of endoleak(NCT01965717) has recently been completed. The

aim of the study was to establish the correlation ofMMP-9 with specific types of endoleaks and therequirement for re-intervention. The results havenot been made public yet.

– Endostaples have offered satisfactory results after thecompletion of the pivotal study of the AptusEndovascular AAA Repair System (NCT00507559).The ANCHOR (Aneurysm Treatment Using the Heli-FX Aortic Securement System Global Registry) study iscurrently recruiting patients (NCT01534819) aimingfor a primary completion date in 2020. The primaryoutcome measures are the prevention of graftmigration and the treatment of Type Ia endoleak.

– Navigation systems in CT offer more accurate needleplacement and as the number of direct sac interventionswill increase accurate needle placement under CTfluoroscopy will be necessary. The Endoleak RepairGuided by Navigation Technology study(NCT01843322) is a small study of 27 patients that isrecently completed and is aiming to delineate whetherthe treatment of type II endoleaks can be improved byadding navigation technology in terms of precision andreduction of radiation exposure.

– Novel polymers will be developed after the Nellixsystem, regardless of the fact that the results untiltoday have not been as expected. The novel ANEUFIXsystem in the treatment of endoleaks is assessed in afeasibility study (NCT02487290). The study is a non-randomised, multi-centre safety and feasibility trial ofAneufix ACP-T5 to treat patients with isolated type IIendoleaks in the presence of a non-shrinking AAA sacfollowing an EVAR procedure; however, it has onlyrecruited 4 patients at the moment.

ConclusionWe may conclude that as the treatment options forendovascular repair of abdominal aortic aneurysms andthe complexity of devices increase, there will be anincreased necessity of prevention and management ofendoleaks. Radiology is crucial in the management ofsuch a phenomenon and needs to offer a number ofsolutions in the endoleak prevention and management.

Fig. 6 a-c CTA scan showing continuous expansion of the aneurysmal sac after initial repair for rupture. The expansion occurred over three yearsreaching a size of nearly 10 cm but without any evidence of an endoleak. It was considered as a result of “endotension”

Orgera et al. Insights into Imaging (2019) 10:91 Page 8 of 10

AbbreviationsCT: Computed Tomography; DSA: Digital subtraction angiography;DSPE: Direct sac puncture embolisation; DUS: Doppler ultrasound;ePTFE: Expanded poly-tetrafluoroethylene; EVAR: Endovascular aortic repair;EVAS: Endovascular aneurysm sealing; FEVAR: Fenestrated endovascular aorticrepair; IMA: Inferior mesenteric artery; IVC: Inferior vena cava; MRI: Magneticresonance imaging; TAE: Transarterial embolisation

Authors’ contributionsGO design of the manuscript and data collection, AMT design of themanuscript and data collection, FL, data collection, PL, data collection, ARdesign of the manuscript and data collection, IP design of the manuscriptand data collection, MR manuscript overview, MK manuscript overview andsubmission. All authors read and approved the final manuscript.

Competing interestsThe authors declare that they have no competing interests.

Author details1Department of Radiology, Sant’ Andrea University Hospital La Sapienza,Rome, Italy. 2Department of Radiological Sciences, Sapienza University ofRome, Rome, Italy. 3The Department of Surgical and Biomedical Sciences,University of Perugia, Perugia, Italy. 4The Department of Radiology, AberdeenRoyal Infirmary, NHS Grampian, Aberdeen, UK. 5Department of Radiology,Cambridge University Hospitals NHS Foundation Trust, Hills Road, CambridgeCB2 0QQ, UK.

Received: 14 December 2018 Accepted: 22 July 2019

References1. Parodi JC, Palmaz JC, Barone HD (1991) Transfemoral intraluminal graft

implantation for abdominal aortic aneurysms. Ann Vasc Surg 5:491–4992. White GH, Yu WJ, May J, Chaufour X, Stephen MS (1997) Endoleak as a

complication of endoluminal grafting of abdominal aortic aneurysms:classification, incidence, diagnosis, and management. J Endovasc Surg 4:152–168

3. Orandi BJ, Dimick JB, Deeb GM, Patel HJ, Upchurch GR Jr (2009) Apopulation-based analysis of endovascular versus open thoracic aorticaneurysm repair. J Vasc Surg 49:1112–1116

4. Drury D, Michaels JA, Jones L, Ayiku L (2005) Systematic review of recentevidence for the safety and efficacy of elective endovascular repair in themanagement of infrarenal abdominal aortic aneurysm. Br J Surg 92:937–946

5. Prinssen M, Buskens E, de Jong SE et al (2007) Cost-effectiveness ofconventional and endovascular repair of abdominal aortic aneurysms:results of a randomized trial. J Vasc Surg 46:883–890

6. Lederle FA, Freischlag JA, Kyriakides TC et al (2009) Open VersusEndovascular Repair (OVER) Veterans Affairs Cooperative Study Group.Outcomes following endovascular vs open repair of abdominal aorticaneurysm: a randomized trial. JAMA 302(14):1535–1542

7. Powell JT, Sweeting MJ, Ulug P et al (2017) Meta-analysis of individual-patient data from EVAR-1, DREAM, OVER and ACE trials comparingoutcomes of endovascular or open repair for abdominal aortic aneurysmover 5 years. Br J Surg. 104(3):166–178

8. Paravastu SC, Jayarajasingam R, Cottam R, Palfreyman SJ, Michaels JA,Thomas SM (2014) Endovascular repair of abdominal aortic aneurysm.Cochrane Database Syst Rev 23:CD004178

9. Patel R, Sweeting MJ, Powell JT, Greenhalgh RM, EVAR trial investigators(2016) Endovascular versus open repair of abdominal aortic aneurysm in 15-years’ follow-up of the UK endovascular aneurysm repair trial 1 (EVAR trial1): a randomised controlled trial. Lancet 388(10058):2366–2374

10. Kouvelos GN, Oikonomou K, Antoniou GA, Verhoeven EL, Katsargyris A(2017) A systematic review of proximal neck dilatation after endovascularrepair for abdominal aortic aneurysm. J Endovasc Ther. 24(1):59–67

11. Antoniou GA, Smyth JV, Antoniou SA, Serracino-Inglott F, Murray D (2012)Endoleak is the Achilles’ heel of the chimney technique for the treatment ofcomplex aortic disease. Int Angiol 31(6):595–596

12. Massimi TM, Kostun ZW, Woo EY (2017) Transcaval embolization of a type Igutter endoleak after three-vessel chimney endovascular aneurysm repair. JVasc Surg. 65(5):1515–1517

13. Zaiem F, Almasri J, Tello M, Prokop LJ, Chaikof EL, Murad MH (2018) Asystematic review of surveillance after endovascular aortic repair. J VascSurg 67(1):320-331.e37

14. Ullery BW, Tran K, Itoga NK, Dalman RL, Lee JT (2017) Natural history ofgutter-related type Ia endoleaks after snorkel/chimney endovascularaneurysm repair. J Vasc Surg. 65(4):981–990

15. Quinn AA, Mehta M, Teymouri MJ et al (2017) The incidence and fate ofendoleaks vary between ruptured and elective endovascular abdominalaortic aneurysm repair. J Vasc Surg. 65(6):1617–1624

16. Maleux G, Poorteman L, Laenen A et al (2017) Incidence, etiology, andmanagement of type III endoleak after endovascular aortic repair. J VascSurg. 66(4):1056–1064

17. Dudeck O (2013) Endoleaks - when is treatment necessary? [Article inGerman]. Radiologe. 53(6):526–530

18. Sidloff DA, Gokani V, Stather PW, Choke E, Bown MJ, Sayers RD (2014) TypeII endoleak: conservative management is a safe strategy. Eur J VascEndovascular Surg 48(4):391–399

19. Kuziez MS, Sanchez LA, Zayed MA (2016) Abdominal aortic aneurysm type IIendoleaks. J Cardiovasc Dis Diagn 4:5

20. El Batti S, Cochennec F, Roudot-Thoraval F, Becquemin JP (2013) Type IIendoleaks after endovascular repair of abdominal aortic aneurysm are notalways a benign condition. J Vasc Surg 57(5):1291–1297

21. Cieri E, De Rango P, Isernia G et al (2015) Type II endoleak is an enigmaticand unpredictable marker of worse outcome after endovascular aneurysmrepair. J Vasc Surg. 59(4):930–937

22. Lo RC, Buck DB, Herrmann J et al (2016) Risk factors and consequences ofpersistent type II endoleaks. J Vasc Surg. 63(4):895–901

23. Moulakakis KG, Klonaris C, Kakisis J et al (2016) Treatment of type II endoleakand aneurysm expansion after EVAR. Ann Vasc Surg [Epub ahead of print]

24. Mangialardi N, Orrico M, Ronchey S, Praquin B, Alberti V, Setacci C (2016)Towards an entirely endovascular aortic world: an update of techniques andoutcomes for endovascular and open treatment of type I, II, and IIIendoleaks. J Cardiovasc Surg (Torino). 57(5):698–711

25. Abdulrasak M, Resch T, Sonesson B, Holst J, Kristmundsson T, Dias NV (2017)The long-term durability of intra-operatively placed Palmaz stents for thetreatment of type Ia endoleaks after EVAR of abdominal aortic aneurysm.Eur J Vasc Endovascular Surg 53(1):69–76

26. Schlösser FJV, de Vries JPPM, Chaudhuri A (2017) Is it time to insertendoanchors into routine EVAR? Eur J Vasc Endovascular Surg 53(4):458–459

27. Montelione N, Pecoraro F, Puippe G et al (2015) A 12-year experience withchimney and periscope grafts for treatment of type I endoleaks. J EndovascTher. 22(4):568–574

28. Bianchini Massoni C, Mascoli C, Perini P et al (2018) Endovascular treatmentsfor type Ib endoleaks after aorto-iliac aneurysms exclusion: mid-term results.Int Angiol. 37(5):384–389

29. Naughton PA, Garcia-Toca M, Rodriguez HE et al (2011) Endovascular treatment ofdelayed type 1 and 3 endoleaks. Cardiovasc Intervent Radiol. 34(4):751–757

30. Ward TJ, Cohen S, Fischman AM et al (2013) Preoperative inferiormesenteric artery embolization before endovascular aneurysm repair:decreased incidence of type II endoleak and aneurysm sac enlargementwith 24-month follow-up. J Vasc Interv Radiol. 24(1):49–55

31. Alerci M, Giamboni A, Wyttenbach R et al (2013) Endovascular abdominal aneurysmrepair and impact of systematic preoperative embolization of collateral arteries:endoleak analysis and long-term follow-up. J Endovasc Ther. 20(5):663–671

32. Piazza M, Frigatti P, Scrivere P et al (2013) Role of aneurysm sacembolization during endovascular aneurysm repair in the prevention oftype II endoleak-related complications. J Vasc Surg. 57(4):934–941

33. Biancari F, Mäkelä J, Juvonen T, Venermo M (2015) Is inferior mesentericartery embolization indicated prior to endovascular repair of abdominalaortic aneurysm? Eur J Vasc Endovascular Surg 50(5):671–674

34. Manunga JM, Cragg A, Garberich R et al (2017) Preoperative inferiormesenteric artery embolization: a valid method to reduce the rate of type IIendoleak after EVAR? Ann Vasc Surg 39:40-47.

35. Quinones-Baldrich W, Levin ES, Lew W, Barleben A (2014) Intraproceduraland postprocedural perigraft arterial sac embolization (PASE) for endoleaktreatment. J Vasc Surg. 59(2):538–541

36. Zanchetta M, Faresin F, Pedon L, Ronsivalle S (2007) Intraoperative intrasacthrombin injection to prevent type II endoleak after endovascularabdominal aortic aneurysm repair. J Endovasc Ther. 14(2):176–183

37. Muthu C, Maani J, Plank LD, Holden A, Hill A (2007) Strategies to reduce therate of type II endoleaks: routine intraoperative embolization of the inferior

Orgera et al. Insights into Imaging (2019) 10:91 Page 9 of 10

mesenteric artery and thrombin injection into the aneurysm sac. J EndovascTher. 14(5):661–668

38. Samura M, Morikage N, Mizoguchi T et al (2018) Effectiveness ofembolization of inferior mesenteric artery to prevent type II endoleakfollowing endovascular aneurysm repair: a review of the literature. Ann VascDis 11(3):259–264

39. Holden A (2014) Endovascular aneurysm sealing for abdominal aorticaneurysm repair: evolution or revolution? Cardiovasc Intervent Radiol. 37(5):1129–1136

40. Böckler D, Holden A, Thompson M et al (2015) Multicenter Nellixendovascular aneurysm sealing system experience in aneurysm sac sealing.J Vasc Surg 62:290–298

41. Brownrigg JR, DeBruin JL, Rossi L et al (2015) Endovascular aneurysmsealing for infrarenal abdominal aortic aneurysms: 30-day outcomes of 105patients in a single centre. Eur J Vasc Endovascular Surg 50:157–164

42. Zerwes S, Nurzai Z, Leissner G et al (2016) Early experience with the newendovascular aneurysm sealing system Nellix: first clinical results after 50implantations. Vascular 24:339–347

43. Reijnen MM, de Bruin JL, Mathijssen EG et al (2016) Global experience withthe Nellix endosystem for ruptured and symptomatic abdominal aorticaneurysms. J Endovasc Ther 23:21–28

44. Carpenter JP, Cuff R, Buckley C et al (2016) Results of the Nellix systeminvestigational device exemption pivotal trial for endovascular aneurysmsealing. J Vasc Surg 63:23–31

45. Abularrage CJ, Patel VI, Conrad MF, Schneider EB, Cambria RP, Kwolek CJ (2012)Improved results using Onyx glue for the treatment of persistent type 2endoleak after endovascular aneurysm repair. J Vasc Surg. 56(3):630–636

46. Güntner O, Zeman F, Wohlgemuth WA et al (2014) Inferior mesentericarterial type II endoleaks after endovascular repair of abdominal aorticaneurysm: are they predictable? Radiology 270(3):910–919

47. Ribé L, Bicknell CD, Gibbs RG et al (2017) Long-term results of intra-arterialonyx injection for type II endoleaks following endovascular aneurysm repair.Vascular 25(3):266-271

48. Baum RA, Cope C, Fairman RM, Carpenter JP (2001) Translumbarembolization of type 2 endoleaks after endovascular repair of abdominalaortic aneurysms. J Vasc Interv Radiol 12(1):111–116

49. Stavropoulos SW, Carpenter JP, Fairman RM, Golden MA, Baum RA (2003)Inferior vena cava traversal for translumbar endoleak embolization afterendovascular abdominal aortic aneurysm repair. J Vasc Interv Radiol 14(9, Pt1):1191–1194

50. Van Bindsbergen L, Braak SJ, van Strijen MJ, de Vries JP (2010) Type IIendoleak embolization after endovascular abdominal aortic aneurysm repairwith use of real-time three-dimensional fluoroscopic needle guidance. JVasc Interv Radiol 21(9):1443–1447

51. Chung R, Morgan RA (2015) Type 2 Endoleaks Post-EVAR: Current evidencefor rupture risk, intervention and outcomes of treatment. CardiovascIntervent Radiol. 38(3):507–522

52. Scali ST, Vlada A, Chang CK, Beck AW (2013) Transcaval embolization as analternative technique for the treatment of type II endoleak afterendovascular aortic aneurysm repair. J Vasc Surg 57:869–874

53. Mansueto G, Cenzi D, Scuro A et al (2007) Treatment of type II endoleakwith a transcatheter transcaval approach: results at 1-year follow-up. J VascSurg 45:1120–1127

54. Giles KA, Fillinger MF, De Martino RR, Hoel AW, Powell RJ, Walsh DB (2015)Results of transcaval embolization for sac expansion from type II endoleaksafter endovascular aneurysm repair. J Vasc Surg. 61(5):1129–1136

55. Ogawa Y, Hamaguchi S, Nishimaki H et al (2015) Embolization by directpuncture with a transpedicular approach using an isocenter puncture (ISOP)method in a patient with a type II endoleak after endovascular aortic repair(EVAR). Cardiovasc Intervent Radiol. 38(3):731–735

56. Uthoff H, Katzen BT, Gandhi R, Peña CS, Benenati JF, Geisbüsch P (2012)Direct percutaneous sac injection for postoperative endoleak treatmentafter endovascular aortic aneurysm repair. J Vasc Surg 56:965–972

57. Yang RY, Tan KT, Beecroft JR, Rajan DK, Jaskolka JD (2017) Direct sacpuncture versus transarterial embolization of type II endoleaks: anevaluation and comparison of outcomes. Vascular 25(3):227-233

58. Chaikof EL, Blankensteijn JD, Harris PL et al (2002) Reporting standards forendovascular aortic aneurysm repair. J Vasc Surg 35:1048–1060

59. Verma H, Meda N, Tripathi RK (2014) Management of complex type Iiiaendoleak by brachio-femoral realignment and interposition stent graft andreview of the literature. Indian J Vasc Endovascular Surg 1:29–32

60. Oderich GS (2011) Reporting on fenestrated endografts: surrogates foroutcomes and implications of aneurysm classification, type of repair, andthe evolving technique. J Endovasc Ther 18:154–156

61. Faries PL, Cadot H, Agarwal G, Kent KC, Hollier LH, Marin ML (2003)Management of endoleak after endovascular aneurysm repair: cuffs, coils,and conversion. J Vasc Surg. 37:1155–1161

62. Espinosa G, Ribeiro Alves M, Ferreira Caramalho M, Dzieciuchowicz L, SantosSR (2009) A 10-year single-center prospective study of endovascularabdominal aortic aneurysm repair with the talent stent-graft. J EndovascTher. 16:125–135

63. Espinosa G, Ribeiro M, Riguetti C, Caramalho MF, Mendes WD, Santos SR(2005) Six-year experience with talent stent-graft repair of abdominal aorticaneurysms. J Endovasc Ther. 12:35–45

64. Forbes TL, Harrington DM, Harris JR, DeRose G (2012) Late conversionof endovascular to open repair of abdominal aortic aneurysms. Can JSurg. 55:254–258

65. Klonaris C, Lioudaki S, Katsargyris A et al (2014) Late open conversion afterfailed endovascular aortic aneurysm repair. J Vasc Surg. 59(2):291–297

66. Kouvelos G, Koutsoumpelis A, Lazaris A, Matsagkas M (2015) Late openconversion after endovascular abdominal aortic aneurysm repair. J VascSurg. 61(5):1350–1356

67. Gilling-Smith G, Brennan J, Harris P, Bakran A, Gould D, McWilliams R (1999)Endotension after endovascular aneurysm repair: definition, classification,and strategies for surveillance and intervention. J Endovasc Surg 6:305e7

68. Trocciola SM, Dayal R, Chaer RA et al (2006) The development ofendotension is associated with increased transmission of pressure andserous components in porous expanded polytetra-fluoroethylene stent-grafts: characterization using a canine model. J Vasc Surg 43(1):109–116

69. Cho J-S, Park T, Kim JY, Chaer RA, Rhee RY, Makaroun MS (2010) Priorendovascular abdominal aneurysm repair provides no survival benefitswhen the aneurysm ruptures. J Vasc Surg 52:1127e34

70. Kougias P, Lin PH, Dardik A, Lee WA, El Sayed HF, Zhou W (2007) Successfultreatment of endotension and aneurysm sac enlargement withendovascular stent graft reinforcement. J Vasc Surg 46:124e7

71. Moll FL, Powell JT, Fraedrich G et al (2011) Management of abdominalaortic aneurysms clinical practice guidelines of the European Society forVascular Surgery. Eur J Vasc Endovascular Surg 41(Suppl 1):S1e58

72. Koole D, Moll FL, Buth J et al (2011) Annual rupture risk of abdominal aorticaneurysm enlargement without detectable endoleak after endovascularabdominal aortic repair. J Vasc Surg 54:1614e22

73. Chandarana H, Godoy MC, Vlahos I et al (2008) Abdominal aorta: evaluationwith dual-source dual-energy multidetector CT after endovascular repair ofaneurysms – initial observations. Radiology. 249(2):692–700

74. Nevala T, Biancari F, Manninen H et al (2010) Type II endoleak afterendovascular repair of abdominal aortic aneurysm: effectiveness ofembolization. Cardiovasc Intervent Radiol 33(2):278–284

75. Massis K, Carson WG 3rd, Rozas A, Patel V, Zwiebel B (2012) Treatment oftype II endoleaks with ethylene-vinyl-alcoholcopolymer (Onyx). VascEndovascular Surg 46(3):251–257

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Orgera et al. Insights into Imaging (2019) 10:91 Page 10 of 10